Microfluidic device for on-chip mixing and encapsulation ... · experiments in double emulsion drops to test the influence of the DNA concentration on the amount of green fluorescence

1

Microfluidicdeviceforon-chipmixingandencapsulationoflysates

Jui-ChiaChang,aZoeSwank,bOliverKeiser,aSebastianMaerkl,b*EstherAmstada*

a.InstituteofMaterials,EcolePolytechniqueFédéraledeLausanne(EPFL),Lausanneb.InstituteofBioengineering,EcolePolytechniqueFédéraledeLausanne(EPFL),

Lausanne

Abstract

Emulsion drops are often employed as picoliter-sized containers to perform screeningassays. These assays usually entail the formation of drops encompassing discrete objectssuch as cells or microparticles and reagents to study interactions between the differentencapsulants. Drops are also used to screen influences of reagent concentrations on thefinal product. However, these latter assays are less frequently performed because it isdifficulttochangethereagentconcentrationoverawiderangewithhighprecisionwithinasingle experiment. In this paper, we present a microfluidic double emulsion dropmakercontainingpneumaticvalvesthatenableinjectionofdifferentreagentsusingpulsedwidthmodulationandsubsequentmixing.Thisdevicecanproducedropsfromreagentvolumesaslow as 10 µl with minimal sample loss, thereby enabling experiments that would beprohibitivelyexpensiveusingdropletgeneratorsthatdonotcontainvalves.Weemploythisdevice to monitor the kinetics of cell free synthesis of green fluorescent proteins insidedoubleemulsions. Todemonstrate thepotentialof thisdevice,weperformDNA titrationexperimentsindoubleemulsiondropstotesttheinfluenceoftheDNAconcentrationontheamountofgreenfluorescenceproteinsproduced.

Introduction

Emulsion drops are well-suited containers for performing chemical and biochemicalreactionsunderwell-definedconditionsand involumes thatare significantly smaller thanthoserequiredtoconductreactionsinbulk.Thisisespeciallybeneficialforhighthroughputscreeningassays.1-3Theaccuracyofsuchassaysdependsonthedegreeofcontroloverthecompositionandconcentrationofreagentscontainedinsidethedropsaswellastheirsizedistribution.Dropswith a narrow size distribution canbeproducedusingmicrofluidics.4-6Thesedropshave,forexample,beenemployedascontainersfordrugscreeningassays,7,8toperformpolymerasechainreactions(PCR)fromviruses,9,10orsinglecells,11fordirectedevolution of enzymes,12 to study the secretion of proteins on a single cell level,13 or toidentify genes that are responsible for a certain cellular phenotype.14 To perform thesescreeningassays,reagentsareoftenpre-mixedbeforetheyareinjectedintothedevice.Pre-mixinglimitskineticstudiestocharacterizingveryslowreactionsorthelatestagesoffasterreactionsbecausereagentsstarttoreactbeforetheyareloadedintodrops.Moreover,pre-mixingprevents in situ changesof the relative reagent concentrations such thatonlyone

.CC-BY-NC-ND 4.0 International licensenot peer-reviewed) is the author/funder. It is made available under aThe copyright holder for this preprint (which was. http://dx.doi.org/10.1101/247627doi: bioRxiv preprint first posted online Jan. 14, 2018;

http://dx.doi.org/10.1101/247627

http://creativecommons.org/licenses/by-nc-nd/4.0/

2

solution composition can be screened per experiment. A possibility to overcome theseshortcomings is the injection of reagents into drops after they have been formed forexamplethroughtheapplicationofhighelectricfields15-19ortheadditionofchemicalsthatdestabilize drops.20 However, the number of different reagents that can be controllablyaddedtointactdropsislimited.Moreover,itisdifficulttoaccuratelyandcontinuouslyvarytheconcentrationof injectedreagents.Apossibility tograduallyandcontrollablyvary thereagent concentration is to co-flow two fluids under laminar conditions; in this case, thereagentexchangeisdiffusionlimited.21-23However,becausemixingreliesondiffusion,thespatio-temporalcontroloverthesolutioncompositionispoor.Thiscontrolcanbeimprovedif mixing is enhanced, for example by introducing turbulences into the fluid flow usingstructured microchannels24, 25 or active mixers, such as micropumps, or micromixers.26However,italwaystakessometimetoequilibrateinjectionflowratesespeciallyifmultiplefluids are involved. Thus, even with mixing features being implemented, it is difficult tocontrollably and continuously change the concentrations of different reagents with hightemporalresolution.

The concentrationof reagents canbe changedoverawide rangeandonvery short timescalesifmicrofluidicchannelsareequippedwithpneumaticvalvesthatcanbeopenedandclosedrapidly.Thesevalvesallowtheformationofatrainofalternatingplugsofdifferenttypesofmiscibleliquidsthatcanbesubsequentlymixed.Thelengthoftheplugofeachfluidscales with the duty cycle, corresponding to the fraction of time one valve is openedcompared to the entire cycle time, and can be adjusted in situ. This procedure enablesvarying relative reagent concentrations over a wide range within a single experiment bygradually changing the duty cycles; thereby only minimal volumes of reagents areconsumed. This method, the so-called pulsed width modulation (PWM), has beenimplementedinmicrofluidicdevices27,28andisoftenemployedtosynthesizebiopolymers,to study the influence of their composition on their function, and to test the effect ofcertainmoleculesonthecellbehavior.29-34Recentadvancementsofthistechnologyenableindependent injectionofup tosixdifferent reagentsandchanging theirconcentrations insitu by up to five orders ofmagnitudes.35 This level of compositional control over such awideconcentrationrangeisdifficulttoachievewithco-flowingfluids.DespitethesedistinctadvantagesofPWM,on-chipmixingofsolutionsthataresubsequentlyprocessedintodropsofdefinedsizesisthusfardonethroughco-flowofdifferentfluids.Theabilitytorapidlyandcontrollablychangetheconcentrationofreagentsoveramuchwiderrangewouldopenupnew possibilities for high throughput screening assays. However, pneumatic valves havenever been implemented into microfluidic flow focusing drop makers and consequentlyPWMhasthusfarnotbeenperformedinthesedevices.

Inthispaper,wepresentamicrofluidicflowfocusingdropmakerthathasthree inletsforreagents, each of them controlled by a pneumatic valve. This device allows separateinjection of different reagents using PWM, on-chip mixing, and formation of doubleemulsiondropsofdefineddimensionsusingthismixtureasaninnerphase.Thepneumaticvalvesprovideanadditionalbenefit: theyenableencapsulationof liquidswithvolumesaslow as 10 µl at an efficiency approaching 100%. This device is employed to encapsulatelysates that synthesize green fluorescentprotein (GFP) insidedoubleemulsiondrops.We


http://dx.doi.org/10.1101/247627


3

demonstratethepotentialofthedevicetoperformcharacterizationassaysbytitratingDNAandmeasuringtheinfluenceofitsconcentrationontheamountofGFPproducedindoubleemulsiondrops.

Results

Wefabricatemicrofluidicdevicesusingsoftlithography.36Thesedevicescontainthreeinletsforaqueoussolutions that formthe innerphase,one inlet for theoil thatconstitutes themiddle phase, and one inlet for an aqueous phase that forms the surrounding liquid. Tocontrolthefluidflowinthethreeinletsfortheinnermostphase,weintroducepneumaticvalvesontopofthesechannels,asschematicallyshownbythebluefeaturesinFigure1a.36Thevalvesenableinjectionofdifferentfluidsusingpulsedwidthmodulation,asexemplifiedin the optical micrograph in Figure 1b.35, 37, 38 To accelerate the mixing of the differentinjected reagents, herringbones are introduced to the serpentine-like sectionof themainchannellocatedbetweentheinletsfortheinnermostphasesandthatfortheoilphase,25asschematicallyshowninFigure1a.Toseparatereagent-loadeddropsfromemptyones,thedevicehasaT-junctionusedasasortingunit,asshownschematicallyinFigure1aandintheoptical micrograph in Figure 1c. The flow of double emulsions is again controlled withpneumaticvalvesthatchangethehydrodynamicresistanceoftheoutletchannels.

Figure1: Schematic illustrationof themicrofluidicdevice (a)Overviewof thedevicewiththe inlet for (A) the outermost aqueous phase, (B) the oil phase and (C-E) the threeinnermostaqueousphases.Eachinletfortheinnermostphasecontainsapneumaticvalvethatenables controlling the fluid flow (F-H); thecontrol linesare indicated inblue.Dropsexitthedevicethroughoneofthetwooutletchannelsthatalsocontainpneumaticvalves(I-J). (b)Opticalmicrographof anoperating devicewhere an aqueousphase and a dye arealternatively injected using PWM. (c) Optical micrograph of double emulsions that arecollectedthroughtheleftoutlet.


http://dx.doi.org/10.1101/247627


4

Topreciselytunetheamountofliquidinjectedintothemainchannel,theswitchingtimesofthe valves must be accurately controlled. Valves close if their control channels arepressurizedsuchthatthevalvesarepressedtowardsthebottomofthefluidchannel.Weemploy solenoid valves to pressurize and depressurize the control channels; these valvesare electronically controlled. To determine the response time of the valves, we use acoloredfluidandquantifytheintensityprofileacrossthefluidchannelunderneaththevalveasafunctionofthepulsewidth.Ifthevalveisopen,theentireinletchannelunderneaththevalve is colored, as shown in theopticalmicrograph in Figure 2a; in our device, the inletchannelsectionbelowthevalve is200µmwide. If thevalve isclosed, the fluid ispushedasidesuchthatthissectionbecomestransparent,asshowninFigure2b.Ifthepulsewidthof the valve, corresponding to the time the valve is open, is 20ms, no fluid canpass, asdemonstrated by the flat curve of the black circles in Figure 2c. Ifwe increase the pulsewidthto30ms,somefluidpassesandevenmorefluidpassesifthepulsewidthisincreasedto40ms,asindicatedbytheincreasedpeakintensity,shownbythebluesquaresandredstars inFigure2c. Increasingthepulsewidthfurtherbroadenstheintensitypeakbutdoesnotincreaseitsamplitudeanymore,asshowninFigure2c.Theseresultsindicatethattheresponsetimeofthevalves,andthustheirminimumpulsewidth,is40ms.Thispulsewidthsetsalowerlimittothesizeofplugsthatcanbeformed.Thissizedependsonthefluidflowratethatwesetto250µl/h;inthiscase,theminimumplugvolumeis3.5nl.

Figure2:Characterizationofthepneumaticvalves.(a,b)Opticalmicrographsofapneumaticvalvethatis(a)openedand(b)closed.(c)Theintensityofthecolorantmeasuredacrossthechannelunderneaththevalveforpulsewidthsof10ms,(●)for20ms;(ƞ )for25ms;(Ȍ )for30ms;(ƴ )for40ms;(ʆ )for50ms.(d,e)Opticalmicrographsofdropsformedwithapulsewidthof(c)100msand(d)800ms.(f)Thestandarddeviationofthedyeintensityindropsasafunctionofthepulsewidth.


http://dx.doi.org/10.1101/247627


5

A key feature of the device is its ability to precisely change in situ the composition ofreagentsthatareencapsulatedindoubleemulsions.Thisdegreeofcontrolisonlypossibletoobtain,ifadjacentplugscontainingdifferentreagentsarehomogeneouslymixedbeforethe solution is compartmentalized into drops. The longer the plugs are, the longer itwilltaketoconvertanarrayofplugsintoahomogeneoussolutionbecausethediffusionlengthsareincreased.Toquantifythemaximumlengthofplugsthatourdevicecanmixbeforethesolutionisbrokenintodrops,weemploytwoliquids,waterandawater-solubledye,asaninner phase. The lengths of the plugs are varied by changing the pulsewidth of the twocorrespondingvalves.Whentheaqueousphasereachesthewater-oiljunction,itisbrokeninto70µmdiameterdrops. If the twoaqueousphasesare fullymixedwhen the solutionreaches thewater-oil junction, the intensityof thedrops ishomogeneous,as indicated inFigure2d.Bycontrast,iftheplugsaretoolong,thecompositionofthesolutionattheoil-waterjunctionvariesovertimeandtheintensityofthedropsisheterogeneous,asshowninFigure2e.Toquantifythemaximumlengthofaqueousplugsthatcanbefullymixedinourdevice,wemeasurethestandarddeviationofthedropintensityasafunctionofthepulsewidth; these experiments are performed using liquids whose viscosity is close to that ofwater.Ifthepulsewidthisbelow400ms,theintensityofdropsisuniform,asshownintheopticalmicrographinFigure2dandsummarizedinFigure2f.Bycontrast,ifthepulsewidthisincreasedabovethisvalue,theintensitybecomesheterogeneous,asshownintheopticalmicrograph in Figure 2e and summarized in Figure 2f. These results indicate that themaximumpulsewidth is 400ms corresponding to a plug volume of 35 nl. Hence, in ourdevice, we can vary the plug volumes from 3.5 to 35 nl. However, these are notfundamentallimits.Thedynamicrangeofthedevicecouldbeincreasedbyprolongingthemixingunit,orbyalteringthechanneldimensions.

Microfluidics allows encapsulation of reagents with a high efficiency. However, in mostcases, some reagents are lost during device start-up. This fluid loss poses no problem ifreagentsareinexpensiveandavailableinlargequantities.However,somebiologicalassaysinvolve expensive reagents or samples that are only available in very small volumes. Toprocesssmallervolumesinmicrofluidicdevices,samplelossesmustbeminimized.Thiscanbe achieved using pneumatic valves because they allow initialization of the device withwaterpriortoswitchingtotheexpensivereagent.Whenthedevicerunsstably,thechannelforthewaterphaseisclosedwhilechannelscontainingexpensivereagentsareopened.Toseparate reagent-loadeddrops fromemptyones,weagainemploypneumatic valves thatareincorporatedintothecollectionchannels.Totesttheperformanceofthesevalves,weformaqueoussingleemulsiondropsandcollectthemthroughtherightoutlet.Eveniftheright valve is open, 80 µm diameter drops tend to breakwhen they pass it because thechannelheightunderneaththevalvecannotexceed14µmforthemtobeabletofullyclosethe channel. To prevent drop break-up, we increase the height underneath the sortingvalves to 20µm. These valves do not completely close the channel even if they are fullypressurized. Instead, they decrease the height of the channel, thereby increasing itshydrodynamicresistanttosuchanextentthatunderouroperatingconditionsdropsdonotpass this obstacle anymore.We exemplify this behavior by closing the valve of the leftcollectionchannelandopeningtheoneoftherightchannel.Inthiscase,dropsdonotbreak


http://dx.doi.org/10.1101/247627


6

whentheypass theopenvalveandcanbecollectedontherightside,as indicated in theoptical micrograph in Figure S1a. Tomaximize the accuracy of the sorting, we close thevalve of the right channel 10 ms before we open the valve of the left channel: Thisprocedurereducespressuredifferencesbetweenthetwocollectionchannelsandthereforeallows thedrops to flow into thedesiredcollectionchannelassoonas thecorrespondingvalveisopened.Iftheinjectionrateisbelow600µl/h,dropscanbesortedwithoutanyloss,asshownbytheopticaltime-lapseimagesinFigureS1a.Bycontrast,ifthisinjectionrateisexceeded, drops tend to split at the sorting junction during the switching operation, asshownintheopticaltimelapsimagesinFigureS1b.

Fordropstobeusedasreactionvessels,theymustbestableduringincubation.Totestthestabilityofdrops,we load themwith lysates39, 40anda solutioncontaining30mM3-PGAand incubate them at 29°C for 3 h. Unfortunately, most of the single emulsion dropscoalesce during their incubation such that the resulting sample is polydisperse. Thiscoalescence limits the usability of drops formany screening applications. The stability ofdropscoatedwithperfluorinatedtriblockcopolymerscanbecompromisedbythepresenceofhighconcentrationsof ions.41Totest ifwecan increasethedropstability,wedecreasetheconcentrationof3-PGAto4mM.However,alsointhiscase,singleemulsiondropstendto coalesce albeit to a smaller extent, as summarized in Figure 3a and shown in thefluorescence micrographs in Figures S2a and S2b. We expect the coalescence of singleemulsiondropstobecausedbyionsthatareincloseproximitytothesurfactantsortodropinterfaces. If this expectation holds, double emulsions should be more stable againstcoalescence because their outer liquid-liquid interface is separated from ions by the oilshell.Indeed,thepercentageofintactdoubleemulsionsissignificantlyhigherthanthatofsingleemulsiondrops,evenifasolutioncontaining30mM3-PGAisusedasaninnermostphase,assummarizedinFigure3a.

TotestifwecanexpressGFPindoubleemulsions,weemployPURE,acell-freetranscription/ translation reaction mixture generated from purified components.42 PURE is nowcommerciallyavailable(NEB)butisarelativelyexpensivereagent;itcurrentlycostsover1USDper µl. Therefore, it is beneficial to only consume small volumes. Small volumes aredifficulttohandlewithstandardmicrofluidicdeviceswherevolumesbetween50and100µlcan easily be lost during start-up. To prevent sample loss during start-up of our device,deionizedwateris injectedthroughinletC.Whenthedevicerunsstably,valveFisclosed,anaqueoussolutioncontainingPURE is injectedthrough inletD,andanaqueoussolutioncontaining DNA is injected through inlet E. The two reagent-containing solutions areinjectedusingPWMwithpulsewidthsof50ms.Theaqueousmixtureisbrokeninto65µmdiameter double emulsion drops that display a narrow size distribution. These drops areincubated at 29°C for 3 h and the formation kinetics of functional GFP is measured byacquiringafluorescentmicrographevery8min.Within2hfluorescencereachesaplateauasshowninthefluorescenceintensitytraceinFigure3b.TheseexperimentsshowfeasibilitytosynthesizeGFPindoubleemulsionsandtheeconomicuseofexpensivereagentssuchasPURE.


http://dx.doi.org/10.1101/247627


7

Figure 3: Synthesis of GFP in drops. (a) The percentage of drops that remain intact as afunctionoftheincubationtimeforsingleemulsiondropsloadedwith30mM(Ƶ )and4mM3-PGA(ƴ )anddoubleemulsiondropscontaining30mM(○)and4mM3-PGA(●).(b)ThenormalizedfluorescenceintensitymeasuredindoubleemulsiondropsloadedwithPUREasa function of the incubation time. (c) The normalized fluorescence intensity of lysatesolutions in bulk (ʆ ), in single emulsiondrops loadedwith lysates throughPWM (ƴ ) andwithpre-mixed lysates (ƴ ),anddoubleemulsiondrops loadedwith lysatesthroughPWM(●)andwithpre-mixedlysates(●)asafunctionoftheincubationtime.Asacontrol,lysateextractwasloadedintodropsintheabsenceofDNA(ƞ ).Doubleemulsionsaresignificantlymorestablethansingleemulsions.Nevertheless,alargefractionofthecorescontaining30mM3-PGAandlysateextractedfromE.colimergewiththe surrounding aqueous phase such that all the encapsulants are released, as shown inFigure3a.To increase the fractionof intactdoubleemulsiondrops,we reduce the3-PGAconcentration to 4mM. This reduction in 3-PGA concentration significantly increases thepercentageof intact double emulsions, as exemplified in the fluorescencemicrographs inFiguresS2c,dandsummarized inFigure3a.Hence, solutionscontaining4mM3-PGAareemployedforthefollowingexperiments.

To investigate the kinetics and amount of synthesized functional GFP, we quantify thefluorescence intensity of each drop. Encapsulation of lysates into 65 µm diameter dropsdoes not alter the kinetics of GFP synthesis, as summarized in Figure 3c. However, theamountofGFPproduced indoubleemulsions is significantly lower than thatproduced insingleemulsiondropsand inbulk, as indicatedby thebluediamond in Figures3c. Inourexperiments, single emulsion drops encompass 0.55 µM functional GFP whereas doubleemulsion drops only contain 0.39 µM functional GFP, as summarized in Figure S3. Thisdifferencemight be caused by a partial leakage of reagents through the shell of doubleemulsions,byanalogytothecrosstalkobservedbetweenaqueousdropsthataredispersedin perfluorinated oils.43 Nevertheless, there is a significant amount of functional GFPsynthesizedindoubleemulsiondrops,indicatingthattheycanbeusedtoscreeneffectsofsynthesisconditionsontheamountoffunctionalGFPproduced.

The main feature of our microfluidic device is its ability to change the reagentconcentrationsinsituwithhighaccuracy.Toexploitthisfeature,weperformDNAtitrationexperiments: Water is injected through inlet D, an aqueous solution containing lysate


http://dx.doi.org/10.1101/247627


8

extractsfromE.coli,30mM3-PGA,andanenergysourcethroughinletC,andanaqueoussolutioncontaining15nMDNAthroughinletE.Thecycletimeofthevalvesiskeptconstantat4swhilethepulsewidthofvalvesGandHarevariedfrom40msto320msinfoursteps.Using this procedure, we screen four DNA concentrations, 13.3, 11.7, 6.7, and 1.7 nM,withinasingleexperimentthatconsumesaslittleas10µloflysatesandsimilarvolumesofthe energy solution. From these reagents,we produce approximately 100 drops for eachDNA concentration.We repeat the same experiment using a solution containing 7.5 nMDNA to screen additional four DNA concentrations. The resulting double emulsions areincubatedfor3hat29°Candtheirintensityisquantifiedusingfluorescentmicroscopy.Theamountofproteinproduced inadrop increaseswith increasingamountsofDNAup toaconcentration of 6 nM and levels off thereafter, as shown by black circles in Figure 4. Asimilar trend is seen in the experiments performed in bulk, as shown by red triangles inFigure4.However,usingdrops,weobtaina100timesimprovedstatisticswhileconsuming8-fold less reagents than if thescreening isperformed inbulk.These resultsdemonstratethepotentialofourdevicetoscreendifferentreactionconditionswithverylowvolumesofreagents.

Figure4: In situDNA-titration. The influenceof theDNA concentrationon the amountofGFPsynthesizedinsidedoubleemulsions(○)andinbulk(Ƶ ).ForGFPsynthesizedindoubleemulsions,theDNAconcentrationwasvariedinsitubypulsewidthmodulation.

Conclusions

Wepresentamicrofluidicdropmakerthatallowsmixingofuptothreeliquidsonchipusingpulsedwidthmodulationbeforetheresultingmixtureisencapsulatedindoubleemulsionsthatdisplayanarrowsizedistribution.Employinggreenfluorescentproteinsynthesizedinacell-freereactionasamodelsystem,wedemonstratethatthisdevicecanbeusedtoscreeninfluencesofthesynthesisconditionsontheamountofproteinproducedwhileconsumingonlysmallvolumesofreagents.Inaddition,thisdevicerequiresminimalmanualoperation,


http://dx.doi.org/10.1101/247627


9

therebyeliminatingtheriskforpipettingerrors.Thesefeaturesareofparticularimportancefor the production of expensive biomolecules and for screening and characterization ofsamples thatareonlyavailable inverysmallquantities.Hence, thisdevicemightopenupnewpossibilitiestoscreensynthesisconditionsalsoforreactionsthatinvolveexpensiveorrarereagents.

ConflictsofinterestTherearenoconflictstodeclare.

AcknowledgementsJ.-C.ChangwasfinanciallysupportedbytheEPFLFoodandNutritioncenterandZ.Swankbythe EuropeanResearchCouncil (ERC) under the EuropeanUnion’sHorizon 2020 researchandinnovationprogram(GrantAgreementNo.723106).

References

1. Y.Ding,J.ChooandA.J.deMello,MicrofluidicsandNanofluidics,2017,21,58.2. M.T.Guo,A.Rotem,J.A.HeymanandD.A.Weitz,LabonaChip,2012,12,2146-

2155.3. H.N.JoenssonandH.A.Svahn,Angew.Chem.Int.Ed.,2012,51,12176-12192.4. S.L.Anna,N.BontouxandH.A.Stone,Appl.Phys.Lett.,2003,82,364-366.5. R.K.Shah,H.C.Shum,A.C.Rowat,D.Lee,J.J.Agresti,A.S.Utada,L.Y.Chu,J.W.

Kim,A.Fernandez-Nieves,C.J.MartinezandD.A.Weitz,Mat.Today,2008,11,18-27.

6. R.Dangla,S.C.KayiandC.N.Baroud,PNAS,2013,110,853-858.7. J.Q.Boedicker, L. Li, T.R.KlineandR. F. Ismagilov,LabonaChip, 2008,8, 1265-

1272.8. O.J.Miller,A.ElHarrak,T.Mangeat,J.C.Baret,L.Frenz,B.ElDebs,E.Mayot,M.L.

Samuels,E.K.Rooney,P.Dieu,M.Galvan,D.R.LinkandA.D.Griffiths,PNAS,2012,109,378-383.

9. M.M.Kiss,L.Ortoleva-Donnelly,N.R.Beer,J.Warner,C.G.Bailey,B.W.Colston,J.M.Rothberg,D.R.LinkandJ.H.Leamon,AnalyticalChemistry,2008,80,8975-8981.

10. N.R.Beer,E.K.Wheeler,L.Lee-Houghton,N.Watkins,S.Nasarabadi,N.Hebert,P.Leung,D.W.Arnold,C.G.BaileyandB.W.Colston,AnalyticalChemistry,2008,80,1854-1858.

11. Y. Zeng, R. Novak, J. Shuga,M. T. Smith and R. A.Mathies,Analytical Chemistry,2010,82,3183-3190.

12. J.J.Agresti,E.Antipov,A.R.Abate,K.Ahn,A.C.Rowat,J.C.Baret,M.Marquez,A.M.Klibanov,A.D.GriffithsandD.A.Weitz,PNAS,2010,107,4004-4009.

13. L.Mazutis,J.Gilbert,W.L.Ung,D.A.Weitz,A.D.GriffithsandJ.A.Heyman,NatureProtocols,2013,8,870-891.

14. B. L. Wang, A. Ghaderi, H. Zhou, J. Agresti, D. A. Weitz, G. R. Fink and G.Stephanopoulos,NatureBiotechnology,2014,32,473-U194.

15. D.J.Eastburn,A.SciambiandA.R.Abate,PlosOne,2013,8.16. M.Zagnoni,C.N.BaroudandJ.M.Cooper,PhysicalReviewE,2009,80.


http://dx.doi.org/10.1101/247627


10

17. H. Gu,M. H. G. Duits and F.Mugele, International Journal ofMolecular Sciences,2011,12,2572-2597.

18. M.Chabert,K.D.DorfmanandJ.L.Viovy,Electrophoresis,2005,26,3706-3715.19. D. R. Link, E.Grasland-Mongrain,A.Duri, F. Sarrazin, Z.D. Cheng,G. Cristobal,M.

MarquezandD.A.Weitz,Angew.Chem.Int.Ed.,2006,45,2556-2560.20. I.Akartuna,D.M.Aubrecht,T.E.KodgerandD.A.Weitz,LabonaChip,2015,15,

1140-1144.21. M.A.Holden, S. Kumar, E. T. Castellana,A. Beskok andP. S. Cremer,Sensors and

ActuatorsB-Chemical,2003,92,199-207.22. C.Neils,Z.Tyree,B.FinlaysonandA.Folch,LabonaChip,2004,4,342-350.23. A.Baccouche,S.Okumura,R.Sieskind,E.Henry,N.Aubert-Kato,N.Bredeche, J.F.

Bartolo, V. Taly, Y. Rondelez, T. Fujii and A. J. Genot,Nature Protocols, 2017, 12,1912-1932.

24. J.Marschewski,S.Jung,P.Ruch,N.Prasad,S.Mazzotti,B.MichelandD.Poulikakos,LabonaChip,2015,15,1923-1933.

25. A.D.Stroock,S.K.W.Dertinger,A.Ajdari,I.Mezic,H.A.StoneandG.M.Whitesides,Science,2002,295,647-651.

26. N.T.NguyenandZ.G.Wu,JournalofMicromechanicsandMicroengineering,2005,15,R1-R16.

27. A. Ainla, I. Gözen, O. Orwar and A. Jesorka,Analytical Chemistry, 2009,81, 5549-5556.

28. D.Irimia,D.A.GebaandM.Toner,AnalyticalChemistry,2006,78,3472-3477.29. X. Zhang,A.Grimley, R. BertramandM.G. Roper,Analytical Chemistry, 2010,82,

6704-6711.30. C.L.Hansen,S.Classen,J.M.BergerandS.R.Quake,JACS,2006,128,3142-3143.31. K.R.King,S.Wang,A.Jayaraman,M.L.YarmushandM.Toner,LabonaChip,2008,

8,107-116.32. B.Krishnaswamy,C.M.Austin,J.P.Bardill,D.Russakow,G.L.Holst,B.K.Hammer,

C.R.ForestandR.Sivakumar,IeeeTransactionsonCommunications,2013,61,5139-5151.

33. Y.Kim,B.Kuczenski,P.R.LeDucandW.C.Messner,LabonaChip,2009,9,2603-2609.

34. Y.Hori,C.Kantak,R.M.MurrayandA.R.Abate,LabonaChip,2017,17,3037-3042.35. K.WoodruffandS.J.Maerkl,AnalyticalChemistry,2017,90,696–701.36. Y.N.XiaandG.M.Whitesides,AnnualReviewofMaterialsScience,1998,28,153-

184.37. M.A.Unger,H.P.Chou,T.Thorsen,A.SchererandS.R.Quake,Science,2000,288,

113-116.38. C.L.Hansen,M.O.A.SommerandS.R.Quake,PNAS,2004,101,14431-14436.39. Z.Z.Sun,C.A.Hayes,J.Shin,F.Caschera,R.M.MurrayandV.Noireaux,Jove-Journal

ofVisualizedExperiments,2013,DOI:10.3791/50762.40. Y.C.KwonandM.C.Jewett,ScientificReports,2015,5.41. G.Etienne,M.KesslerandE.Amstad,MacromolecularChemistryandPhysics,2017,

218.42. Y. Shimizu,A. Inoue, Y. Tomari, T. Suzuki, T. Yokogawa,K.NishikawaandT.Ueda,

NatureBiotechnology,2001,19,751-755.


http://dx.doi.org/10.1101/247627


11

43. P.Gruner,B.Riechers,B. Semin, J. Lim,A. Johnston,K. Short and J. C.Baret,Nat.Comm.,2016,7.


http://dx.doi.org/10.1101/247627


12

SupportingInformation

MaterialsandMethods

Materials

GFP is synthesized by mixing an aqueous solution containing cell-free reagents and anenergy solution. The cell-free reagent solution contains, lysate extracted fromE. coli andGFP DNA templates. The energy solution is composed of water containing 10.5 mMmagnesiumglutamate,100mMpotassiumglutamate,0.25mMdithiothreitol(DTT),1.5mMof each amino acid except leucine, 1.25mM leucine, 50mM HEPES, 1.5mM adenosinetriphosphate (ATP), and guanosine-5'-triphosphate (GTP), 0.9 mM cytidine triphosphate(CTP)anduridinetriphosphate(UTP),0.2mg/mLtRNA,0.26mMcoenzymeA(CoA),0.33mMnicotinamideadeninedinucleotide (NAD),0.75mMcyclicadenosinemonophosphate(cAMP),0.068mMcolonicacid,1mMspermidine,2%PEG-8000,4mM3-Phosphoglycericacid(3-PGA).

FabricationofMicrofluidicDevice.

Themicrofluidicdeviceismadeofpoly(dimethylsiloxane)(PDMS)usingsoftlithography.1,2Itcontainsfiveinlets,onefortheouterphase,oneforthemiddlephase,andthreefortheinnerphases.Inaddition,itcontainsfiveinletsforthecontrolvalveswhereairisinjectedtoclosethepneumaticvalveslocatedontopoftherespectivefluidchannels.Thematersusedfor thebottompartof thedevicethatcontainsthe liquidchannels is fabricatedfromtwolayersofnegativephotoresist,SU-8;thefirstlayeris14-20µmtall,thesecondlayeris100µm tall. Themasters employed to fabricate the top part of the device is made of threelayersofphotoresist:The first layer is14μmtallandcomposedofapositivephotoresist,AZ9260,thesecondandthirdlayersare20µmtallandcomposedofanegativephotoresist,SU-8.

ThemicrofluidicdeviceismadefromSylgard184PDMS(DowCorning).Thethreepartsarejoinedthroughreactivebonding:Tofabricatethetoppartofthedevice,weemployabase:crosslinkerratioof1:5,themiddlepartismadeataratiobase:crosslinkerratioof1:20,andthebottompartataratiobase:crosslinkerratioof1 :10.3Themiddlepartmustbethin to ensure that the valves are sufficiently flexible to close the fluid channels if thecontrolchannelsarepressurized.Tocontrolthethicknessofthemiddlepartofthedevice,wespincoatPDMStoforma100µmthicklayer.PDMSiscuredat80°Cfor25minutes.Thetopandmiddlepartsarealignedandbondedbyincubatingthemat80°Cfor2hours.Theresulting part is removed from the mold and bonded to the bottom part using oxygenplasmafollowedbyincubationat65⁰Cfor12hours.Theresultingdeviceshave100µmtallcontrol channels. The inlet liquid channels are 20 µm tall and lead into the threedimensional junction where the outermost liquid phasemeets themain channel; at thisjunctionthechannelheightisincreasedto60µm.

To producewater-oil-water double emulsions, the part upstream the junctionwhere theoutermost phase flows into the main channel must be hydrophobic whereas the main


http://dx.doi.org/10.1101/247627


13

channel further downstream must be hydrophilic. To render the top part of the devicehydrophobic we inject fluorinated oil (Novec 7500, 3M, MN) containing 1 vol%trichloro1H,1H,2H,2H-perfluorooctyl)silane (Sigma-Aldrich, MO) into this section of thedevice.Torendertheremainingpartofthedevicehydrophilic,thesurfacesaretreatedwithan aqueous solution containing 2 wt% poly(diallyldimethylammonium chloride) and 1 Msodiumchloride.

Fluids are injected into the device through polyethylene tubing (PE/5, ScientificCommoditiesInc.,AZ)usingsyringepumps.

Encapsulationofcell-freereagents

Weemployanaqueoussolutioncontaining10wt%ofpoly(vinylalcohol)(PVA),Mw13000-23 000 Da, 87-89% hydrolyzed, as an outer phase, a perfluorinated oil, HFE7500 (Novec7500,3M,MN),containing1wt%ofsurfactant4,5asamiddlephase,andanaqueousphaseasaninnerphase.Toinitializethedevice,weusedeionizedwaterasaninnerphase.Oncethe device runs stably,we close the valve F that controls the fluid flow of the deionizedwater.Simultaneously,thetwoothervalves,DandE,arealternatinglyopenedtoinjecttheaqueous solutions containing the reagents from the twoother inlets for the innerphase.Within one experiment, we inject 10 µl of an aqueous solution containing lysate extractsolutionthroughinletDand10µlofanaqueoussolutioncontainingDNAandenergysourcethroughinletE.Thereby,wekeepthedutycycleofeachvalveat50ms,resultinginatotalcycletimeat100ms.

DNAtitration

ToperformDNAtitrationexperiments,weemploythreedifferentaqueousphasesasinnerphases:InletEcontainslysatewithanenergysource,inletDcontainsanaqueoussolutionwith 15 nM of DNA, and through inlet C we inject pure water. We change the DNAconcentration, without changing the concentration of any other reagent, by varying thedutycyclesofthevalveGandHthatcontroltheflowoftheaqueoussolutionscontainingpurewaterandDNArespectively.Thedutycyclesare400ms.ToenlargetherangeofDNAconcentrations that are screened,we repeat the same experiment but inject an aqueoussolutioncontaining7.5nMofDNAthroughinletD.Dropsaresubsequentlyincubatedat29°Cfor3h.Duringthisincubation,wemonitortheformationofgreenfluorescenceproteinusingfluorescencemicroscopywhereoneimageisacquiredevery8minutes.

Analysisoftheformationofgreenfluorescenceproteins

The formation of GFP is quantified using fluorescence micrographs. The averagefluorescence intensityofeachdrop isquantifiedandnormalizedfor itssizeandshape.Tocorrectforlensingeffectsthatoccuratthedropinterfaces,weperformcontrolexperimentswhere the fluorescence intensityof solutions containing knownamountsof fluorescein ismeasured inbulk, insingleemulsion,anddoubleemulsiondrops.Lensingeffects increase


http://dx.doi.org/10.1101/247627


14

thefluorescenceintensityinsingleemulsionsby11%,comparedtothebulkandindoubleemulsions by 15%. We correct for these lensing effects and convert the fluorescenceintensityintoaproteinconcentrationusingacalibrationcurvemeasuredinbulk.

Dropsorting

To separate reagent-containing drops from empty ones, we introduce a T-junctiondownstream thedrop generation junction. Bypressurizing the left control channel of thesortingunit,theleftvalvepartiallyclosesthechannelsuchthatitshydrodynamicresistanceincreasesanddropsflowintotherightoutlet.Toswitchthedirectionofthefluidflow,weclosetherightvalve10msbeforeweopenthe leftone. If the fluid flowupstreamtheT-junctiondoesnotexceed600µl/h,dropsfollowthefluidflowandremainintactevenwhenthefluidflowswitchesdirection,asshown inthetime lapseopticalmicrographs inFigureS1a. By contrast, if the injection flow rate exceeds 600 µl/h, some drops split at the T-junctionwhilethedirectionofthefluidflowischanged,resultinginsomelossofreagents,asshowninthetimelapseopticalmicrographsinFigureS1b.

FigureS1:Dropsorting.Timelapseopticalmicroscopyimagesillustratingthechangeintheflowdirectionofdropsifthetotalinjectionspeedis(a)600µl/hand(b)900µl/h

Synthesisofgreenfluorescenceproteinsindrops

To test if we can track the formation of GFP inside drops,we formmonodisperse singleemulsiondropswithadiameterof70µmthatareloadedwithlysatesand4mMof3-PGA.EventhoughthePGAconcentrationinthesedropsisalmostanorderofmagnitudebelowthePGAconcentrationtypicallyused,mostofthedropscoalesce,asindicatedbythehighpolydispersity of drops incubated at 29°C for 30min, shown in Figure S2a and the evenhigherpolydispersityofdropsaftertheyhavebeenincubatedatthistemperaturefor3h,asshown in Figure S2b. Coalescence of drops hampers their use for screening assays. Weexpectthecloseproximityofahighconcentrationofionsclosetotheliquid-liquidinterfaceto deter drop stability. To test this expectation, we produce double emulsion dropscontaining lysatesand4mM3-PGAintheircore;theouter liquid-liquid interfaceofthese


http://dx.doi.org/10.1101/247627


15

drops is separated from ions by their oil shell. Indeed, double emulsions aremuchmorestable against coalescence, as indicated by their narrow size distribution after they havebeenincubatedat29°Cfor30min,asshowninFigureS2cand3h,asshowninFigureS2d.

FigureS2:SynthesisofGFPindropswith4mM3-PGA.(a-d)Fluorescencemicrographsof(a,b) single emulsion and (c, d) double emulsion drops loadedwith lysates an incubated at29°Cfor(a,c)30minand(b,d)3h.

Quantificationofgreenfluorescentproteinconcentrations

ToquantifytheamountofGFPproducedinsingleanddoubleemulsiondropswemeasureacalibrationcurveinbulk.TheamountofGFPproducedinsolutionscontaining4mM3-PGAis approximately 50% lower compared to solutions containing 30mM, as summarized inFigure S3. By contrast, the kinetics of the GFP production is not affected by theconcentrationof3-PGA,asshowninFigure3cinthemainpaper.

FigureS3:ConcentrationofGFPsynthesizedin65μmdropscontaining4mMand30mM3-PGA.GFPissynthesizedinsingleemulsions(blue),indoubleemulsions(green),andinbulksolution(yellow)


http://dx.doi.org/10.1101/247627


16

References1. Y.N.XiaandG.M.Whitesides,AnnualReviewofMaterialsScience,1998,28,153-

184.2. A.Rotem,A.R.Abate,A.S.Utada,V.VanSteijnandD.A.Weitz,LabonaChip,2012,

12,4263-4268.3. C.Holtze,A.C.Rowat, J. J.Agresti, J.B.Hutchison,F.E.Angile,C.H. J.Schmitz,S.

Koster,H.Duan,K. J.Humphry,R.A. Scanga, J. S. Johnson,D.PisignanoandD.A.Weitz,LabonaChip,2008,8,1632-1639.

4. G.Etienne,M.KesslerandE.Amstad,MacromolecularChemistryandPhysics,2017,218.


http://dx.doi.org/10.1101/247627


Documents

Microfluidic device for on-chip mixing and encapsulation ... · experiments in double emulsion drops to test the influence of the DNA concentration on the amount of green fluorescence